摘要
提出了基于神经网络的框架结构损伤多重分步识别方法,建立了用于框架结构损伤识别的高效神经网络。根据构件损伤的多重分步识别思路,把构件损伤识别过程分为:利用神经网络建立损伤异常过滤器对构件损伤进行预警;以频率构造的组合指标作为神经网络输入向量,对构件损伤进行初步定位;以频率和模态振型构造的组合指标作为神经网络输入向量,对构件损伤进行具体定位;以频率平方变化率作为神经网络输入向量,对构件损伤程度进行识别。最后针对三跨四层的框架结构进行了损伤识别数值模拟。结果表明:基于神经网络的框架结构损伤多重分步识别方法简化了网络的结构,能够有效地对框架结构损伤进行预警、定位和定量。
The multi-stage damage identification method for frame structures based on neural network was proposed.A kind of high efficient neural network to identify damage in frame structures was established.This method was divided into four steps according to the multi-stage identification ideas of member damage.Firstly,damage anomalous filter was set up by neural network to alarm the damage in structural members.Secondly,the primary location of the member damage was determined by the neural network with inputing the combined damage index of frequency.At the third step,the specific location of the member damage was determined by the neural network with inputing the combined damage index of frequency and vibration mode.Finally,the damage degree of the member was identified by neural network with inputing the change rate of squared modal frequency.Numerical simulation of damage identification for three-span four-floor frame structure was carried out.Results show that multi-stage damage identification method based on neural network simplifies network structure,and can alarm,locate and quantify the damage effectively.
出处
《建筑科学与工程学报》
CAS
2011年第1期106-111,共6页
Journal of Architecture and Civil Engineering
基金
重庆市科技攻关计划项目(2009AB0040)
关键词
框架结构
神经网络
损伤
多重分步识别方法
frame structure
neural network
damage
multi-stage damage identification method
